Yohan Park

Bio

I am an incoming M.S. student at KAIST's Advanced Machine Intelligence (AMI) Lab advised by Prof. Tae-Hyun Oh. I work on computer vision and multimodal learning with a focus on embodied AI. My interests span the perception-to-action stack, emphasizing 3D scene understanding and scene/task representations that support reliable long-horizon planning and execution.

Looking ahead, I aim to build human-like embodied systems that can generate high-level plans from instructions and perception, acquire new atomic skills, refine them at the action level from both observation and interaction using multimodal perceptual signals, and hierarchically compose these skills to execute the plans.

Publications

*: equal contribution. ^†: corresponding author.
C: conference, J: journal, W: workshop, P: preprint, D: Domestic

• Selected
• All

[P1] DarkEQA: Benchmarking Vision-Language Models for Embodied Question Answering in Low-Light Indoor Environments

Yohan Park, Hyunwoo Ha, Wonjun Jo, Tae-Hyun Oh^†

Arxiv'25: arXiv preprint. (Under review)

Paper Project Page

[C1] Reactive Constraint Relaxation for Urban Environment Navigation

Jinwoo Kim*, Keonyoung Koh*, Seungsup Lee, Yohan Park, Daehyung Park^†

RiTA'24: International Conference on Robot Intelligence Technology and Applications.
Best Student Paper Award

Paper Video

[D2] A Depth-Map-Based Semantic Label Transfer System for Embodied Agents Using Scene Memory
(장면 메모리를 활용한 체화 에이전트용 깊이맵 기반 시맨틱 레이블 전이 시스템)

Yohan Park, Hyunwoo Ha, Tae-Hyun Oh^†

IPIU'26: Workshop on Image Processing and Image Understanding.
Best Paper Award (Honorable Mention)

[P1] DarkEQA: Benchmarking Vision-Language Models for Embodied Question Answering in Low-Light Indoor Environments

Yohan Park, Hyunwoo Ha, Wonjun Jo, Tae-Hyun Oh^†

Arxiv'25: arXiv preprint. (Under review)

Paper Project Page

[C1] Reactive Constraint Relaxation for Urban Environment Navigation

Jinwoo Kim*, Keonyoung Koh*, Seungsup Lee, Yohan Park, Daehyung Park^†

RiTA'24: International Conference on Robot Intelligence Technology and Applications.
Best Student Paper Award

Paper Video

[D1] A Study on the Stability of Humanoid Robots with Respect to Foot Structure
(휴머노이드 로봇의 발 구조에 따른 안정도 연구)

Yohan Park, Donglin Choi

KRoC'17: Korea Robotics Society Annual Conference, Future Robotic Scientist session.
Best Student Paper Award

Press

Experience

Projects

• Selected
• All

Graph-Guided LLM Test Generation
Diversity-based Test Generation for LLMs

Whole-Body Imitation for Mobile Manipulation
9-DoF Control via CVAE and Mobile ALOHA

Graph-Guided LLM Test Generation
Diversity-based Test Generation for LLMs

FSM-Based Manipulation
4-DoF Trajectory Control in Real Time

Whole-Body Imitation for Mobile Manipulation
9-DoF Control via CVAE and Mobile ALOHA

Honors and Awards

Education

Full CV in PDF.